DESIGN AND DEVELOPMENT OF SPEECH DATABASE FOR DIGIT AND VOWELS IN SANSKRIT LANGUAGE AND RECOGNITION SYSTEM USING MACHINE LEARNING

The study of machine voice processing is called speech recognition. The synthesis component may be referred to as speech production. Speech recognition is a prominent and active field of study that translates human speech into computer-recognizable words. It often entails pattern extraction from digitized audio samples and representation of those patterns using the proper data model. The content of these patterns is then determined by comparing them to one another using mathematical processes. The focus of this research is speech recognition of isolated words. The purpose of this study is to identify Sanskrit isolated words uttered by various male and female non-native Sanskrit speakers from Maharashtra's Aurangabad District. The MFCC, or Mel Frequency Cepstral Coefficients In order to test the samples, features from and Support Vector Machine (SVM) of the audio signal have been used. Confusion Matrix is used for classification

Перевод пока недоступен